Published October 5, 2018 | Version 1
Dataset Open

IconArt Dataset

  • 1. Télécom Paris


  • 1. Télécom Paris


The dataset IconArt dataset was introduced in the following paper : "Weakly Supervised Object Detection in Artworks" Gonthier et al. ECCV 2018 Workshop Computer Vision for Art Analysis - VISART 2018.

You can also find project page for the paper here.

This dataset contains 5955 images (from WikiCommons) : a train set of 2978 images and a test set of 2977 images (for classification task). 1480 of the 2977 images are annotated with bounding boxes for 7 classes : ‘angel’,‘Child_Jesus’,‘crucifixion_of_Jesus’,‘Mary’,‘nudity’, ‘ruins’,‘Saint_Sebastien’.


The version 2 of the database contains 10 classes with the same images. The classes are ‘angel’,‘beard’,‘capital’,‘Child_Jesus’,‘crucifixion_of_Jesus’,‘Mary’,‘nudity’,‘ruins’,‘Saint_Sebastien’,‘turban’.


This research was supported in part by the "IDI 2017" project funded by the IDEX Paris-Saclay, ANR-11-IDEX-0003-02.


Files (2.8 GB)

Name Size Download all
1.4 GB Preview Download
1.4 GB Preview Download

Additional details


  • Gonthier et al. (2018) Weakly Supervised Object Detection in Artworks arXiv:1810.02569